Exogenously Triggered Perceptual Switches in Multistable Structure-From-Motion Occur in the Absence of Visual Awareness

Journal of Vision (2016) 16(3):14, 1–16 1

Exogenously triggered perceptual switches in multistable structure-from-motion occur in the absence of visual awareness Department of General Psychology and Methodology, Otto-Friedrich-Universitat¨ Bamberg, Bamberg, Germany Center for Behavioral Brain Sciences, Magdeburg, Germany Cognitive Biology, Otto-von-Guericke Universitat,¨ # Alexander Pastukhov Magdeburg, Germany $

Berlin School of Mind and Brain, Humboldt-Universitat¨ zu Berlin, Berlin, Germany Center for Behavioral Brain Sciences, Magdeburg, Germany Cognitive Biology, Otto-von-Guericke Universitat,¨ Jan-Nikolas Klanke Magdeburg, Germany $

Here, we characterize the duration of exogenously Introduction triggered perceptual switches in an ambiguously rotating structure-from-motion display and demonstrate their independence on visual awareness. To this end, we Typically, we experience our perception as stable and triggered a perceptual reversal by inverting the on-screen unambiguous, in a sense that the same retinal input motion and systematically varied the posttrigger results in the same perception that remains constant presentation duration, while collecting observers’ reports even during prolonged viewing. However, this seeming about the initial and final directions of illusory rotation. one-to-one relationship between sensory inputs and We demonstrate that for the structure-from-motion perception is an illusion (Gregory, 2009; Metzger, display, perceptual transitions are extremely brief (20 2009). This is particularly clear when it is violated by ms) and can be considered instantaneous from an so-called multistable displays that are compatible with experimental perspective. We also report that although several distinct and comparably plausible perceptual very brief posttrigger intervals (10–20 ms) reliably initiate interpretations. These displays force the visual percep- a perceptual reversal, observers become aware of tion to continuously switch between alternatives despite perceptual switches only if the posttrigger presentation constant sensory evidence (Blake & Logothetis, 2002; continues for at least 80 ms. Additional experiments Leopold & Logothetis, 1999). demonstrated that an observed lack of visual awareness The single most studied aspect of multistable for brief posttrigger presentation intervals cannot be attributed to either a systematic delay of visual awareness perception is perceptual switching, and we have a fair, or to backward masking. Our results show that although hardly complete, understanding of how the exogenously triggered perceptual reversal can occur in the occurrence of perceptual reversals can be predicted absence of visual awareness, extending earlier work on from the stimulus properties (Brouwer & van Ee, 2006; spontaneous reversals that indicated that neither Kang, 2009; Levelt, 1965) and prior perceptual awareness nor attention may be required for multistable experience (Blake, Westendorf, & Fox, 1990; Kang & perception. Methodologically, the brevity and the short Blake, 2010; Nawrot & Blake, 1989; Pastukhov & latency of induced perceptual reversals make them Braun, 2011; Wolfe, 1984). The neural correlates of particularly suitable for finely timed experiments, such as endogenous triggers of spontaneous reversals are magneto/electroencephalography studies. currently debated, but recent evidence from imaging

Citation: Pastukhov, A., & Klanke, J. N. (2016). Exogenously triggered perceptual switches in multistable structure-from-motion occur in the absence of visual awareness. Journal of Vision, 16(3):14, 1–16, doi:10.1167/16.3.14.

doi: 10.1167/16.3.14 Received September 14, 2015; published February 12, 2016 ISSN 1534-7362

Downloaded from jov.arvojournals.orgThis work ison licensed 09/23/2021 under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. Journal of Vision (2016) 16(3):14, 1–16 Pastukhov & Klanke 2

words, exogenously triggered perceptual reversals occur in the absence of visual awareness.

Methods

Observers

Procedures were in accordance with the Declaration of Helsinki and were approved by the medical ethics board of the Otto-von-Guericke Universita¨t, Magde- burg: ‘‘Ethik-Komission der Otto-von-Guericke-Uni- versita¨t an der Medizinischen Fakulta¨t.’’ All participants had normal or corrected-to-normal vision. Apart from the second author, observers were naive to the purpose of experiments and were paid for their participation. Figure 1. An exogenously triggered reversal of illusory rotation in structure-from-motion displays. Either an inversion of the 2D motion at time Ttrigger (A, illustrated for two example dots) may Apparatus trigger a reversal of the illusory rotation (B) or the illusory rotation may remain stable, following a spatial readjustment of Stimuli were generated with MATLAB using the individual flow elements (C). See also Movie 1. Psychophysics Toolbox (Brainard, 1997) and displayed on a CRT screen (Iiyama VisionMaster Pro 514, studies suggests that they are localized in sensory areas iiyama.com, resolution 1,600 3 1,200 pixels, refresh of the brain rather than in regions associated with rate 100 Hz). The viewing distance was 73 cm so that executive control and attention (Fra¨ssle, Sommer, each pixel subtended approximately 0.0198. Observers Jansen, Naber, & Einhaüser, 2014; Knapen, Brascamp, responded using a keyboard. Background luminance Pearson, van Ee, & Blake, 2011; Weilnhammer, was kept at 36 cd/m2. The experimental room was lit Ludwig, Hesselmann, & Sterzer, 2013). Less is known dimly (ambient luminance at 80 cd/m2). about the duration of perceptual reversals and about the exact temporal relationship between a trigger event, changes within a sensory representation, and the Display following visual awareness of that switch. This is primarily because we infer the timing of perceptual The SFM stimulus consisted of 50 dots distributed reversals from observers’ immediate responses, which over the surface of the sphere. The sphere diameter was are too variable to provide a reliable estimate 5.78, and the dot diameter was 0.0578. For the main (Pastukhov, Vonau, & Braun, 2012). object (presented during the main Ton interval), the To overcome this limitation, we investigated the dots were distributed in such a way as to ensure a temporal characteristics of exogenously triggered specific distance between all left- and right-moving dots switches (Pastukhov et al., 2012; Treue, Andersen, at the time of the on-screen motion inversion (Ttrigger, Ando, & Hildreth, 1995; see Figure 1; Movie 1). To offset of Tpre/onset of Tpost presentation intervals) to quantify the duration of exogenously triggered per- maximize the probability of triggering a perceptual ceptual switches in structure-from-motion (SFM) switch (see Stonkute, Braun, & Pastukhov, 2012, for displays, we established the duration of the intermedi- details). For the probe stimulus (presented during the ate/mixed perception following an exogenous trigger probe interval), the dots were distributed randomly event. For this, we report that perceptual reversals in over the surface of the sphere. Both main and probe SFM are extremely brief. In addition, we combined stimuli were generated anew on every trial. several experimental measures to dissociate a dominance change within sensory representations and the visual awareness of this change. We demonstrate that Experiment 1 although the inversion of the on-screen motion appears to trigger the reversal of perceptual dominance even if Nine observers (five of them female, four male) the posttrigger presentation is stopped after 20 ms, the participated in the experiment. Each of the four observers become aware of that only if the following experimental conditions (see below) was measured in a presentation period is at least 80 ms long. In other separate experimental session. Each session consisted

Downloaded from jov.arvojournals.org on 09/23/2021 Journal of Vision (2016) 16(3):14, 1–16 Pastukhov & Klanke 3

of eight blocks, and each block contained 70 trials. of the logistic function were obtained using a bootstrap Note that the trials from Experiments 1 and 2 were procedure implemented in the Palamedes toolbox. equally intermixed during each block (i.e., 35 trials belonged to Experiment 1 and 35 to Experiment 2). Each trial consisted of a random onset delay (0.5–1 Experiment 2 s), a pretrigger interval (Tpre ¼ [500, 625, 750, 875, 1000] ms), an optional posttrigger interval (Tpost ¼ [10, 20, 40, Nine observers (five female, four male) participated 80, 160, 320] ms), and a response interval (Figure 2A). in the experiment. The procedure was identical to that The direction of the two-dimensional (2D) motion was of Experiment 1 but for an additional probe SFM inversed at the onset of the posttrigger interval, and the display. The visual sequence of Experiment 1 was presentation continued for a predefined amount of time followed by a brief blank interval (Tblank ¼ 50 ms) and (Tpost). The purpose of the on-screen motion inversion the probe SFM display (Tprobe ¼ 500 ms). The probe was to trigger a reversal of the perceived illusory stimulus was a different sphere (i.e., the location of rotation (Pastukhov et al., 2012; Treue et al., 1995; see individual flow elements was different from that of the Figure 1). The ‘‘no inversion’’ presentation condition main sphere). Observers reported on the initial rotation contained no on-screen motion inversion and, corre- of the main stimulus and on the final direction of spondingly, no postinversion presentation interval illusory rotation of the probe display. See Figure 3A. (Figure 2B). Note that trials from Experiments 1 and 2 were equally The dots were distributed on the surface of the intermixed during each block (i.e., 35 trials belonged to illusory sphere in such a way as to ensure a specific Experiment 1 and 35 to Experiment 2). minimal distance between pairs of left- and right- moving dots at the time of the on-screen motion inversion (Stonkute et al., 2012). We used four interpair Experiment 3 distances to systematically manipulate the strength of Six observers (three of female, three male) partici- the motion transient and, therefore, the probability of pated in the experiment. The SFM display was identical successfully induced perceptual reversals. The four to the Strong (1)/S1 condition of Experiments 1 and 2. conditions were labeled according to the strength of the The presentation schedule of the SFM display was motion transient: Strong (1)/S1, Strong (2)/S2, similar to that of Experiments 1 and 2 but without the Medium/M, and Weak/W. The maximal induced ‘‘no inversion’’ condition and with only long post- destabilization was determined using the longest posttrigger intervals (Tpost ¼ [140, 150, 160, 170, 180] ms). trigger interval duration (Tpost ¼ 320 ms; for further The SFM display was accompanied by a yellow dot details, see the Results sections for Experiments 1 and (diameter 0.758) that moved clockwise along the 2). Please note that the effectiveness of the motion circular trajectory (radius 5.78) with a speed of 6008/s. transient in inducing a perceptual reversal depends not The initial position of the dot was randomized. Please only on its strength but also on a prior perceptual see Movie 2, which demonstrates several presentation history (Pastukhov, Vivian-Griffiths, & Braun, 2015). trials without a response interval (please note that the The same procedure but using variable pretrigger actual experimental display looked different because of intervals has been used to study the onset perception of a higher refresh rate). After the presentation, the SFM SFM displays (Pastukhov, 2015). display was taken of the screen, and the dot was moved Observers reported on initial (beginning of Tpre, to a random location that was at least 458 away from labeled as R(1)) and final (end of Tpost interval or end the location the dot was in at the time of the on-screen of Tpre interval for ‘‘no inversion’’ condition, labeled as motion reversal. The observers were instructed to R(2)) directions of illusory rotation. Observers had an memorize the location of the yellow dot at the time of option of reporting an unclear/mixed percept via the the illusory rotation reversal. They used arrow keys ‘‘down’’ key. The response times were measured with (left and right) to move the dot to the memorized respect to the end of the presentation (offset of Tpost or location and ‘‘Enter’’ to confirm it. Observers had an Tpre for ‘‘no inversion’’ condition). Accordingly, option to report the lack of reversal using a ‘‘Q’’ key perceptual destabilization due to an endogenous trigger (11.6% 6 12.7% of trials). event was quantified as

Preversal ¼ P Rð1Þ 6¼ Rð2Þ : ð1Þ Experiment 4

Group averages were ﬁtted with a logistic function Seven observers (four female, three male) partici- using the Palamedes toolbox (Prins & Kingdom, 2009). pated in the experiment. Apart from an added 500-ms Standard errors of measurement for the free parameters delay before the response prompt, the procedure was

Downloaded from jov.arvojournals.org on 09/23/2021 Journal of Vision (2016) 16(3):14, 1–16 Pastukhov & Klanke 4

Figure 2. Experiment 1. (A) Schematic procedure. Each trial consisted of a random-onset delay (Tdelay ¼ 0.5–1 s), a presentation interval (Ton ¼ Tpre þ Tpost), and a response interval. The direction of the on-screen motion was inversed at the end of the Tpre interval, and the presentation continued for a predefined amount of time (Tpost). The purpose of the on-screen motion inversion was to trigger a reversal of the perceived illusory rotation. Observers reported on the initial (R1) and final (R2) directions of illusory rotation. (B) Schematic procedure for the ‘‘no inversion’’ presentation condition. The procedure was similar but for the omitted on-screen motion

inversion and the lack of the posttrigger interval. (C) Probability of reversal as a function of the posttrigger interval duration Tpost (mean and 95% confidence interval based on binomial distribution). Curves depict best-fitting logistic functions. (D) Distributions of

Downloaded from jov.arvojournals.org on 09/23/2021 Journal of Vision (2016) 16(3):14, 1–16 Pastukhov & Klanke 5

threshold (a) and slope (b) parameters obtained by parametric bootstrapping (1,000 iterations). Solid curves encircle 68.27% of the data points. Bars depict the mean and standard error for each distribution and parameter. (E) Normalized response time for the initial direction of rotation (RT1) as a function of the posttrigger interval duration (Tpost). Gray stripe: SEM for an overall average. (F) Fraction of ‘‘unclear’’ responses for the final direction of rotation (R2) as a function of the posttrigger interval duration (Tpost). Gray stripe: SEM for an overall average.

Figure 3. Experiment 2. (A) Schematic procedure. The procedure was similar to that of Experiment 1 but for a blank interval and a probe display, which followed the presentation of the main stimulus (Tblank ¼ 50 ms, Tprobe ¼ 500 ms, both marked by orange color). The probe display was a different ambiguously rotating SFM sphere. Observers reported first on the initial direction of rotation of the main stimulus R1 and then on the final direction of rotation of the probe stimulus R2. (B–E) Probability of a perceptual switch as the function of the postinversion interval Tpost. Filled circles: results for Experiment 2. Open circles: results for Experiment 1 replotted for comparison. Asterisks mark statistically significant differences between the two experiments (paired-sample t test, a Bonferroni correction for multiple testing). Subfigures show data for (B) Strong (1), (C) Strong (2), (D) Medium, and (E) Weak conditions.

Downloaded from jov.arvojournals.org on 09/23/2021 Journal of Vision (2016) 16(3):14, 1–16 Pastukhov & Klanke 6

Strong0 Strong Medium Weak

Threshold [ms] 64.2 6 10.5 72.6 6 10.6 75.1 6 10.7 76.0 6 11.0 Slope 2.61 6 0.29 2.41 6 0.26 2.45 6 0.30 2.73 6 0.53 Support [ms] 115 156.3 128.3 90.7 Guess rate 0.026 6 0.01 0.028 6 0.01 0.007 6 1.92 0.014 6 0.10 Lapse rate 0.14 6 0.02 0.12 6 0.03 0.25 6 0.24 0.56 6 0.03 Table 1. Experiment 1, summary of logistic function fits.

identical to that of Experiments 1 and 2. Both types of inversion’’) ¼ 0.01 [0.005–0.3], and Preversal (W, ‘‘no trials (with and without probe stimulus) were randomly inversion’’) ¼ 0.02 [0.01–0.04]. mixed within a block. Conversely, Tpost ¼ 320 ms was the longest presentation interval, which provided observers with the best opportunity to observe and report a reversal of illusory Experiment 5 rotation (Tpost ¼ 320 ms in Figure 2C). In agreement with prior work (Stonkute et al., 2012), a stronger Nine observers (five female, four male) participated motion transient due to the on-screen motion inversion in the experiment. The procedure was similar to produced more frequent switches of illusory rotation: Experiment 2. Two conditions were used: ‘‘no inver- Preversal (S1, 320 ms) ¼ 0.84 [0.8–0.88], Preversal (S2, 320 sion’’ and Tpost ¼ 20 ms, labeled here as ‘‘with ms) ¼ 0.85 [0.81–0.89), Preversal (M, 320 ms) ¼ 0.74 [0.7– inversion.’’ The blank duration was systematically 0.79], and Preversal (W, 320 ms) ¼ 0.42 [0.37–0.48]. varied: Tblank ¼ [50, 100, 200, 400, 800] ms. The time intervals in between these two extremes represent a growing probability of reported perceptual switches (Figure 2C). Group averages across the nine observers were fitted with a logistic function. For all Results four conditions, the posttrigger duration that led to threshold reports of visual awareness of perceptual reversals was approximately 65 to 75 ms. Perceptual Experiment 1: Time necessary for the visual switches were reliably (Preversal 0.99Preversal [320 ms]) awareness of the perceptual reversal reported 120 to 150 ms after the trigger event (see Table 1; Figure 2D). In the first experiment, we sought to estimate the Because perceptual adaptation has a strong influence time interval between the trigger event and the display on the perception of multi-stable displays (Blake et al., offset that is necessary for a perceptual reversal and/or 1990; Kang & Blake, 2010; Pastukhov & Braun, 2011; for the visual awareness of it. To this end, we reversed van Ee, 2009), we analyzed its effect on induced the on-screen motion of all flow elements at a perceptual reversals in the current study. First, we predefined moment of time (Ttrigger in Figure 2A), while examined the effect of the short-term adaptation by systematically varying the duration of a posttrigger comparing the probability of reversals for the shortest interval. Observers reported on the initial and final (Tpre ¼ 500 ms) and the longest (Tpre ¼ 1000 ms) directions of illusory rotation for each presentation pretrigger intervals but found no significant change, interval. t(251) ¼1.6, p ¼ 0.1, paired-sample t test. Next, we To confirm that the observers faithfully and consis- repeated the same analysis but for trials from the first tently reported their subjective perception of illusory half versus trials from the second half of each rotation, our experimental design contained two experimental session, to assess the influence of the long- control conditions: ‘‘no inversion’’ and Tpost ¼ 320 ms. term adaptation. Here, we found a small but significant The ‘‘no inversion’’ condition is simply a brief and effect of adaptation, t(251) ¼3, p ¼ 0.003. However, unperturbed presentation of an SFM display (Figure long-term adaptation influenced only an overall prob- 2B), which should lead to stable illusory rotation within ability of induced perceptual reversals (i.e., guess and the presentation interval. Conforming our expecta- lapse rate) but not the threshold or the slope of the tions, the observers tended to report the same direction psychometric functions (data not shown; the effect was of rotation at the beginning and at the end of the weaker than, but qualitatively similar to, the one presentation (No inversion in Figure 2C): Preversal (S1, illustrated in Figure 4). Accordingly, we found no ‘‘no inversion’’) ¼ 0.06 [0.03–0.08] (mean and 95% evidence that, for the paradigm used here, either short- confidence interval for binomial distribution), Preversal term or long-term adaptation alters the speed of (S2, ‘‘no inversion’’) ¼ 0.03 [0.01–0.05], Preversal (M, ‘‘no induced perceptual reversals (see also Experiment 2).

Downloaded from jov.arvojournals.org on 09/23/2021 Journal of Vision (2016) 16(3):14, 1–16 Pastukhov & Klanke 7

Figure 4. Experiment 2, effect of the Tpre interval duration. (A) Probability of the perceptual switch as a function of the posttrigger Tpost interval duration. (B–E) Comparison between the shortest (Tpre ¼ 500 ms, downward-pointing triangles) and longest (Tpre ¼ 1,000 ms, upward- pointing triangles) preinversion intervals. Subfigures show data for (B) Strong (1), (C) Strong (2), (D) Medium, and (E) Weak conditions.

As an additional measure, we have analyzed the effect and therefore, we concentrated on RT1 in the analysis of both condition and posttrigger interval duration on below. With respect to mixed perception responses, we response time and the fraction of mixed reports (see found a marginally significant effect of the posttrigger Table 2). The response times for both intervals (marked interval and a significant interaction of the effects for the as RT1 and RT2 in Figure 2A) were affected by the second response interval, when the observers responded condition and the posttrigger interval duration. How- about the final direction of illusory rotation. The ever, their effect was bigger on the first response interval, weakness of both effects is likely to be explained by a

Downloaded from jov.arvojournals.org on 09/23/2021 Journal of Vision (2016) 16(3):14, 1–16 Pastukhov & Klanke 8

Factor inversion: RT1(‘‘no inversion’’) ¼ 562.6 6 87 ms (mean 6 standard deviation), RT (20 ms) ¼ 536.2 6 67 ms, Condition 1 t(35) ¼ 2.7, p ¼ 0.01 (paired-samples t test). In addition, Condition Tpost 3 Tpost they reported very few mixed percepts (Figure 2F). (d.f. 3/24) (d.f. 6/48) (d.f. 18/144) Thus, all three measures (subjective reports of clear FpFp Fp perception, response times, and subjective reports on mixed perception) indicate that there was no perceptual RT1 16.6 ,0.001 11.8 ,0.001 1.4 0.11 difference between trials with a very brief postinterval RT 5.5 0.005 3.8 0.004 0.75 0.75 2 duration (Tpost ¼ [10, 20] ms) and trials without stimulus Unclear perturbation (‘‘no inversion’’). perception R1 1.7 0.2 1.2 0.3 0.9 0.57 Unclear perception R2 2 0.14 2.2 0.059 1.9 0.019 Experiment 2: Probing an interrupted Table 2. Experiment 1, results for the repeated-measures perceptual switch analysis of variance for response time and mixed reports. Notes: Bold font marks statistically significant effects. Results of Experiment 1 demonstrated that a reversal of illusory rotation was perceived only if the display very low overall fraction of mixed reports (1.8% 6 0.9% presentation continued for another 80 to 150 ms after of all final direction reports across all conditions and all the exogenous trigger (the on-screen motion inversion). postinterval durations). More specifically, the threshold The perceptual switch itself was a very brief event, as duration was associated with longer response times manifested by a very low fraction of mixed percepts (Figure 2E; RT1[‘‘no inversion’’] ¼ 562.6 6 87 ms, even for threshold conditions (see Figure 2F). The RT1[80 ms] ¼ 610 6 106 ms), t(35) ¼3.5, p ¼ 0.001, necessity for this prolonged stimulation may come from paired-samples t test, and a higher fraction of mixed two sources. First, this time may be required for a percepts (Figure 2F; mixed[‘‘no inversion’’] ¼ 0.02 computation of an altered on-screen motion, which in [0.017–0.025] [mean and 95% confidence interval for turn triggers a very brief perceptual switch. This would binomial distribution], mixed[80 ms] ¼ 0.06 [0.046– mean that for very brief postinversion intervals (Tpost ¼ 0.072]), t(35) ¼2.4, p ¼ 0.023, paired-samples t test. In [10, 20] ms), the reversal of an illusory rotation was not other words, the observers were slower to respond and perceived, because the sensory representation of were less certain about the final direction of illusory illusory rotation remained stable at that time point, and rotation for threshold duration displays. However, a low a longer posttrigger presentation is required to trigger a number of mixed reports indicates that mixed phases reversal within them. Second, the subjective awareness were very brief and were rarely perceived even under the of a new direction of illusory rotation may have been most favorable threshold conditions (Tpost ¼ 80 ms). impeded by an earlier perception, for example via Next, we quantified how the motion transient’s forward or backward masking (Enns & Di Lollo, 2000). strength altered the speed of induced perceptual In this case, a reversal of illusory rotation within reversals (i.e., threshold and/or slope of a psychometric sensory representations may have occurred soon after function). Comparing the two most different conditions the on-screen motion inversion (e.g., already after 10- (Strong [1] vs. Weak), we found no significant to 40-ms presentation), and the prolonged presentation differences for the guess rate ( p ¼ 0.98, statistical would be required only to overcome masking. comparison using Monte Carlo method by Palamedes To distinguish between these two possibilities, we toolbox) or for the slope ( p ¼ 0.55) parameters. modified Experiment 1 by appending the display However, there was a highly significant difference in the sequence of Experiment 1 with a brief blank (Tblank ¼ 50 lapse rate ( p , 0.001) and a significant difference in the ms) and a probe stimulus (Figure 3A, additional blank threshold parameters ( p ¼ 0.0183; see also Figure 2D). and probe intervals are marked with orange). The probe Thus, not only did a larger pairing distance produce was presented for 500 ms, giving enough time for the more frequent perceptual switches, but those switches observers to become aware of its direction of rotation, were also perceived slightly earlier (see Table 1). rendering forward masking irrelevant. The probe A critical aspect of the data, which would serve for an stimulus was a different sphere (i.e., the location of important comparison in Experiment 2, is the perception individual flow elements was different from that of the of illusory rotation for very brief posttrigger intervals main sphere). This change interrupted the continuity of (Tpost ¼ [10, 20] ms). Not only have the observers the on-screen motion as well as the continuity of 3D consistently reported the same direction of rotation at representations of individual dots. This way, only the the beginning and at the end of the presentation (Figure representation of an interpolated 3D object could 2C), but they were also very fast to respond (Figure 2E), remain stable. Accordingly, we assumed that because of even faster than on trials without an on-screen motion a very brief interruption, the illusory rotation of the

Downloaded from jov.arvojournals.org on 09/23/2021 Journal of Vision (2016) 16(3):14, 1–16 Pastukhov & Klanke 9

interpolated object should persist (Pastukhov & Braun, 4.6, p , 0.001, were highly significant. However, in 2013) and the direction of rotation of the probe will be both cases, adaptation shifted the entire psychometric representative of the final direction of illusory rotation curve vertically but not horizontally (see Figure 4). As of the main (original) sphere (see Experiment 5 for a with Experiment 1, this indicates that although confirmation of this assumption). In all other respects, accumulated adaptation significantly increases the the procedure was identical to that of Experiment 1. To probability of endogenously triggered perceptual re- facilitate the comparison, trials from Experiments 1 and versals, it has little or no influence on their duration. 2 were mixed together within a single block of an In contrast to Experiment 1, we were unable to fit experimental session (see Methods for details). group averages with a logistic function. The reason for To summarize, our experimental procedure limited this was that the probability of reversal reached its the time for 2D motion extraction, interrupted persis- maximum already after Tpost ¼ 20 ms for all four tence at the level of individual flow elements, preserved experimental conditions. The probability of reversal was persistence at the level of an interpolated object, and significantly different between the two experiments for all gave enough time for the visual awareness of a new postinversion intervals shorter than 160 ms (paired- direction of illusory rotation to emerge. samples t test with a Bonferroni correction for multiple Two postinversion interval durations were used as a tests; statistical significance is indicated by stars in Figure control: ‘‘no inversion’’ and Tpost ¼ 320 ms. As in 3B though E). Note that in Experiment 1, the same Experiment 1, the longest postinversion interval was posttrigger intervals (Tpost ¼ [10, 20] ms) were perceptu- expected to reveal the highest fraction of successfully ally indistinguishable from the ‘‘no inversion’’ condition. triggered switches, and we found no difference between With respect to the research question we formulated the two experiments (see Tpost ¼ 320 ms in Figure 3B for this experiment, this means that the on-screen through E; all p values .0.05). This means that the motion extraction is complete and the reversal of final direction of illusory rotation of the probe stimulus perceptual dominance within sensory representations is was a reliable indicator of the final direction of illusory initiated after 10 to 20 ms of the posttrigger presenta- rotation of the main sphere (see also Experiment 5). tion. This is particularly evident in the Weak condition For the ‘‘no inversion’’ presentation condition, there (Figure 3E), where the fraction of reported switches was no on-screen motion inversion, so the perceptual reaches the maximal level for the same Weak condition switches were not exogenously triggered during the of Experiment 1 already for the 20-ms posttrigger presentation of the main sphere. Therefore, as in interval and remains at this level for all longer Experiment 1, we were expecting the same perception posttrigger intervals. In other words, our probe of illusory rotation to be reported for both displays. paradigm reveals exactly as many switches for brief However, we found that for all four conditions, the posttrigger intervals of Experiment 2 as for the longest probability of the switch was significantly higher in posttrigger intervals in Experiment 1. Accordingly, in Experiment 2 than in Experiment 1 (all p values below Experiment 1, the lack of visual awareness of the 0.01, paired-samples t test; see ‘‘no inversion’’ in Figure reversal occurs despite a dominance change within the 3B through E: results of Experiment 2 are marked by sensory representation of illusory rotation. filled circles, and results of Experiment 1 are marked by Because for all four conditions the maximal destabilization was reached already after the 20-ms posttrigger open circles and are replotted as a comparison): Preversal (S1, 0 ms) ¼ 0.27 [0.22–0.32] (mean and 95% confidence presentation, we were unable to assess the influence of the motion transient’s strength on the speed (or, interval for binomial distribution), Preversal (S2, 0 ms) ¼ conversely, duration) of induced perceptual reversals. 0.36 [0.3–0.41], Preversal (M, 0 ms) ¼ 0.24 [0.2–0.3], and This indicates that a small but significant change in the Preversal (W, 0 ms) ¼ 0.18 [0.14–0.23]. This mild perceptual destabilization is typical for briefly inter- threshold between the Strong (1) and Weak conditions, rupted multistable displays (Kornmeier, Ehm, Bigalke, observed in Experiment 1, most likely was not due to a & Bach, 2007; Orbach, Ehrlich, & Heath, 1963; faster perceptual reversal within a sensory representa- Pastukhov & Braun, 2013) and is likely to reflect an tion. Instead, a stronger motion transient might have accumulated perceptual adaptation/fatigue. To confirm facilitated a faster propagation of this reversal into this and analogously to the analysis we performed for visual awareness, perhaps, by better attracting attention. Experiment 1, we assessed the effects of the short-term and long-term adaptation (respectively, the effect of the preinversion interval duration Tpre and the difference Experiment 3: Estimated time of visual between trials from the first half of an experimental awareness of the illusory rotation reversal session vs. trials from the second half). The effect of during prolonged presentation both short-term adaptation, t(251) ¼10.1, p , 0.001, paired-sample t test for Tpre ¼ 500 ms versus Tpre ¼ 1000 Although a change in the perceptual dominance ms (see Figure 4), and long-term adaptation, t(251) ¼ within the sensory representations is initiated shortly

Downloaded from jov.arvojournals.org on 09/23/2021 Journal of Vision (2016) 16(3):14, 1–16 Pastukhov & Klanke 10

after the trigger event (see Experiment 2), it is possible change and its visual awareness is masking (Enns & Di that the visual awareness of that change consistently Lollo, 2000). For example, the earlier perception of an lags in time (Libet, 1999). In other words, the visual opposite direction of motion could produce forward awareness of an exogenously triggered reversal consis- masking. Alternatively, it is possible that the response tently occurs not earlier than at least 40 to 60 ms after prompt, which appeared on the screen immediately the trigger event. Therefore, we asked observer to after the main display, produced backward masking. estimate the time when they perceived the reversal and Here, we tested the latter hypothesis by replicating examined whether these estimates significantly and the ‘‘Strong (1)’’ condition from Experiment 1 and 2 consistently lagged behind the exogenous trigger event. (marked, respectively, with open and filled circles in To this end, we adopted a ‘‘Libet’s dot’’ paradigm Figure 6) but with an additional 500-ms blank interval (Libet, Gleason, Wright, & Pearl, 1983) that was inserted before the response prompt. To replicate previously used to estimate the time of spontaneous Experiment 1, it was presented after the main SFM reversals of illusory rotation (Pastukhov et al., 2012). display (see also Figure 2A), and the stimulus onset An ambiguously rotating sphere was accompanied by a asynchrony was increased from 20 to 320 ms to 520 to yellow dot that was circling around the SFM display. 820 ms. For the replication of Experiment 2, a blank The observers were instructed to memorize the location was inserted after the presentation of the probe display of the dot at the time when they perceived a reversal in (see also Figure 3A). the illusory rotation (see Movie 2). During the later The results of Experiment 4 are presented in Figure response interval, they moved the dot to the memorized 6. If a delay of visual awareness of the perceptual location, thus allowing us to estimate the time of the reversal in Experiment 1 was due to backward masking, perceived reversal. The initial location of the yellow dot the additional blank should have attenuated its effect. was randomized, and its location at any given time was Therefore, the number of reported perceptual reversals not informative about the time of the trigger event. for brief Tpost durations for the replication of Empirical cumulative densities functions of the Experiment 1 should have increased, and the filled- estimated time of induced perceptual reversals for six circles curve in Figure 6 should have become similar to observers are plotted in Figure 5 (black curves and left the open-circles curve (curves correspond, respectively, y-axis; mean estimated event times relative to the to replications of Experiments 1 and 2). However, the trigger event are marked by teal color). Although the curves in Figure 6 both qualitatively and quantitatively observers varied in the mean estimated time of the match the results of Experiments 1 and 2. Specifically, induced perceptual reversal, they showed no systematic in replication of Experiment 1, short postinversion bias. The group average estimated time of the reversal intervals (T , 40 ms) lead to consistent reports of was 1.5 6 21 ms, and it was not significantly different post perceptual stability. We conclude that the lack of from zero, t(5) ¼ 0.07, p ¼ 0.94. In other words, we awareness is not explained by backward masking from found no tendency to perceive induced perceptual reversals as occurring significantly later than the the response prompt. physical trigger event. Importantly, the same six observers required at least 80 ms of continued visual presentation to develop a visual awareness of an Experiment 5: The direction of illusory rotation induced switch in Experiment 1 (see red lines and right in the probe display reflects the most recent y-axis in Figure 5 replotted for comparison purposes). perceptual state before the interruption Therefore, we conclude that the results of Experiments 1 and 2 cannot be explained by a systematic delay The experimental procedure for the Experiment 2 between the time of the exogenous trigger event and the was based on the assumption that the direction of subjective perception of an illusory rotation reversal. illusory rotation of the probe displays reflects the most recent (final) state of the main display before the interruption. This assumption was based on earlier Experiment 4: Lack of awareness is not due to work that showed that for brief blank intervals, the backward masking perceptual dominance is stabilized by neural persistence—a quickly decaying activity of an originally Our results of Experiments 1 and 2 show that dominant neural population (Pastukhov & Braun, although the switch in perceptual dominance appears 2013). However, it is possible that the perceptual to be initiated if the presentation continues for 20 ms dominance of the probe display reflected another after the on-screen motion inversion (Experiment 2), history effect, such as a sensory memory of multistable the brief presentation times preclude the visual displays (Adams, 1954; Leopold, Wilke, Maier, & awareness of that switch (Experiment 1). One possible Logothetis, 2002; Orbach et al., 1963; Ramachandran explanation for this dissociation between a sensory & Anstis, 1983).

Downloaded from jov.arvojournals.org on 09/23/2021 Journal of Vision (2016) 16(3):14, 1–16 Pastukhov & Klanke 11

Figure 5. Estimated time of an illusory rotation reversal, individual observers. Black color and left vertical axis. Empirical cumulative densities function of an estimated time when a perceptual reversal occurred, relative to the time of the trigger event (Ttrigger). CDF ¼ 0.5 corresponds to the mean estimated time of a perceptual switch (marked by teal color). Although the accuracy and the bias of an estimated switch time vary between individual observers, we found no systematic tendency to perceive an exogenously triggered switch to occur significantly later than the trigger event. Red color and right vertical axis. Results of the Strong (1) condition of Experiment 1 replotted for comparison. In contrast to Experiment 3, all six observers were very similar in that they required .80 ms of continued visual presentation to become aware of an induced perceptual reversal.

To control for this possibility, we replicated Exper- (Tblank ¼ [50, 100, 200, 400, 800] ms). The purpose of iment 2 using only two conditions: ‘‘no inversion’’ and the latter was to dissociate the inﬂuence of two history Tpost ¼ 20 ms (labeled here as ‘‘with inversion’’) but effects in question. Whereas neural persistence decays with a broad range of the blank interval durations within 400 to 500 ms (Pastukhov & Braun, 2013),

Downloaded from jov.arvojournals.org on 09/23/2021 Journal of Vision (2016) 16(3):14, 1–16 Pastukhov & Klanke 12

Figure 7. Experiment 5. Probability of the perceptual switch as a function of the blank interval for trials with (open gray circles) Figure 6. Experiment 4. An additional blank before the response and without (filled orange circles) an on-screen motion interval minimizes the masking but has no effect on perceptual inversion. The influence of the on-screen motion inversion was reversals. Probability of a perceptual switch is plotted as a significant only for very short blank intervals (Tblank , 200 ms). Asterisks mark statistically significant differences between function of the postinversion interval Tpost. Filled circles: replication of Experiment 2. Open circles: replication of ‘‘with inversion’’ and ‘‘no inversion’’ conditions (paired-samples Experiment 1. Asterisks mark the statistically significant t test, the Bonferroni correction for multiple testing). differences between the two conditions (paired-sample t test, the Bonferroni correction for multiple testing). Discussion sensory memory is characterized by very long decay Here, we investigated the perception of an exoge- times of dozens of seconds or even minutes (Leopold et nously triggered reversal of illusory rotation in SFM al., 2002). Accordingly, if the perceptual dominance of displays. We report that the reversals themselves are the probe displays was determined by sensory memory, very brief (Experiment 1) and that the change in the blank interval duration should have a minimal dominance in the sensory representation is initiated effect. Conversely, if illusory rotation was stabilized by shortly after the trigger event, as even a 20-ms neural persistence, this effect should be absent for posttrigger presentation duration is sufficient for this blank intervals longer than 400 to 500 ms. (Experiment 2). However, the observers become aware The results of Experiment 5 are plotted in Figure 7. of that switch only if the presentation continues for at The ‘‘no inversion’’ condition (gray filled circles in least 80 ms after the trigger event (Experiment 1). This Figure 7) served as a baseline to determine the effect of effect cannot be explained either by a systematic delay the blank interval duration on the probability of of visual awareness (Experiment 3) or by backward perceptual reversals in the absence of exogenously masking (Experiment 4). Therefore, we conclude that triggered perceptual switches. The results revealed an exogenously triggered reversals are brief and can occur inverted U-shape and were qualitatively consistent with in the absence of visual awareness. previous reports (Kornmeier et al., 2007; Orbach, Ehrlich, & Vainstein, 1963; Pastukhov & Braun, 2013). For the ‘‘with inversion’’ condition (orange open circles Induced perceptual switches occur in the in Figure 7), the Tblank ¼ 50 ms duration was identical absence of visual awareness to that of Experiment 2 and replicated a reliable switching effect of the prior on-screen motion inver- The results of Experiments 1 and 2 demonstrate that sion. Critically, this effect disappeared for blanks although a reversal of the perceptual dominance within longer than 100 ms. This short-lived effect is consistent the sensory representations of illusory rotation is with neural persistence/hysteresis but not with sensory initiated within 20 ms after the trigger event, the memory of multistable displays or any other bias that observers become aware of that only if the presentation operates at the time scale of seconds. We conclude that continues for at least 80 ms. If the presentation is the perceptual dominance of the probe displays reflects curtailed using shorter posttrigger intervals (Tpost ¼ 10– the latest perceptual state of a prior display. 20 ms), the observers fail to notice the reversal, and

Downloaded from jov.arvojournals.org on 09/23/2021 Journal of Vision (2016) 16(3):14, 1–16 Pastukhov & Klanke 13

both their responses and, presumably, perception are perceptual reversals is clear, it is less obvious how qualitatively and quantitatively similar to that of an individual endogenous triggers differ from exogenous unperturbed stable illusory rotation (see Experiment 1). ones in terms of events occurring at the level of neural Earlier work showed that exogenously triggered representations. reversals also occur in the (near) absence of attention For example, although a multistable display itself (Stonkute et al., 2012). Therefore, we can conclude that may remain constant throughout the entire presenta- neither awareness nor attention is necessary for tion, its retinal image never does. Even when the exogenously triggered reversals of perceptual domi- observers are faithfully fixating, the retinal image of a nance in SFM. display is constantly changing because of eye tremor, Our results raise further questions about the drift, and microsaccades (Martinez-Conde, Macknik, & contribution of top-down factors such as attention and Hubel, 2004). These changes of the retinal image are visual awareness to perceptual switches. Although endogenously generated but can trigger a perceptual shifts of attention were postulated as a possible reversal (van Dam & van Ee, 2006), just like exogenous mechanism behind perceptual reversals (Leopold & changes due to an inversion of the on-screen motion or Logothetis, 1999) and their causal effect on multistable due to a brief change of an image contrast (Kim, perception is well documented (Brouwer & van Ee, Grabowecky, & Suzuki, 2006). 2006; Chong, Tadin, & Blake, 2005; Mitchell, Stoner, & Furthermore, all neural representations, starting Reynolds, 2004), later work demonstrated that atten- already at the retinal level, are intrinsically noisy tion may not be required for spontaneous reversals (Faisal, Selen, & Wolpert, 2008). This means that all (Pastukhov & Braun, 2007; Roeber, Veser, Schro¨ger, & neural representations involved in multistable percep- O’Shea, 2011; but see Brascamp & Blake, 2012; Zhang, tion, both ones that can be considered ‘‘inputs’’ in Jamison, Engel, He, & He, 2011). Similarly, not only modeling terms and those that correspond to compet- exogenously triggered but also endogenously triggered ing percepts, undergo constant random changes. These (spontaneous) reversals can occur without visual noise-driven fluctuations in neural representations are awareness (Brascamp, Blake, & Knapen, 2015; Plato- currently thought to be the main source of spontaneous nov & Goossens, 2014). reversals based on both experimental (Brascamp et al., The similarity between our results and the findings 2006; Pastukhov & Braun, 2011; van Ee, 2009) and on endogenously triggered reversals indicates that modeling (Moreno-Bote, Rinzel, & Rubin, 2007; Noest neural populations responsible for initiation of spon- et al., 2007; Pastukhov et al., 2013) perspectives. Just as taneous perceptual switches are likely to be located in an inversion of the on-screen motion, these noise- sensory regions of the brain. This idea fits well with induced fluctuations are transient and do not produce a prior psychophysical experiments (Alais, Cass, O’Shea, long-lasting bias in favor of a particular perception. & Blake, 2010; Brascamp, van Ee, Noest, Jacobs, & van And, similar to the on-screen motion inversions, they den Berg, 2006; van Ee, 2009), modeling (Noest, van may trigger a reversal in the perceptual dominance Ee, Nijs, & van Wezel, 2007; Shpiro, Moreno-Bote, when they occur at an appropriate moment (Moreno- Rubin, & Rinzel, 2009), and recent imaging studies Bote et al., 2007; Noest et al., 2007). However, just like (Brascamp et al., 2015; Fra¨ssle et al., 2014; Knapen et exogenous triggers, they may also be ignored by the al., 2011). Accordingly, it strengthens the idea that visual system, manifesting themselves as brief periods although both awareness and attention modulate of destabilization (Brascamp et al., 2006; Naber, multistable perception and the occurrence of perceptual Fra¨ssle, & Einhaüser, 2011; Pastukhov & Braun, 2011). reversals, they are not causally responsible for them. Taken together, this suggests that the difference between the endogenous noise-driven transient changes in neural representations and the exogenous-driven Exogenously versus endogenously triggered transient changes in neural representations may be of a perceptual reversals quantitative rather than qualitative nature. Accord- ingly, the interpretation of our results and of other The presented study used exogenously triggered work on the exogenously triggered reversals will be perceptual reversals, and this warrants a question facilitated primarily by a better understanding of about how much our findings can tell us about general divergent endogenous causes of spontaneous reversals. mechanisms behind multistable perception and spontaneous perceptual switches. However, one must remember that a spontaneous reversal can be triggered Duration of perceptual switches by internal forces as divergent as an involuntary eye movement, an intrinsic neural noise, or a shift of Experiment 1 demonstrated that, in agreement with attention. Accordingly, even if a conceptual difference current models of multistable perception (Laing & between exogenously and endogenously triggered Chow, 2002; Moreno-Bote et al., 2007; Moreno-Bote,

Downloaded from jov.arvojournals.org on 09/23/2021 Journal of Vision (2016) 16(3):14, 1–16 Pastukhov & Klanke 14 Knill, & Pouget, 2011; Noest et al., 2007), the duration Acknowledgments of perceptual switches in SFM is extremely brief: The observers rarely reported unclear perception even under the most favorable threshold conditions (80 ms in Experiment 1). These nearly instantaneous switches Commercial relationships: none. may appear to be drastically faster than longer and Corresponding author: Alexander Pastukhov. easily noticeable transitions between two clear percepts Email: [email protected] in binocular rivalry (Blake, O’Shea, & Mueller, 1992; Address: Department of General Psychology and Brascamp et al., 2006; Pastukhov & Braun, 2011). Methodology, Bamberg, Germany. However, it is possible that the difference lies not in the nature of two multistable displays but in the exact deﬁnitions of perceptual switches and perceptual References transitions. For binocular rivalry, a perceptual transition can be Adams, P. A. (1954). The effect of past experience on deﬁned as a perception that is different from the state the perspective reversal of a tridimensional figure. of exclusive visibility and typically includes piecemeal American Journal of Psychology, 67, 708–710, doi: rivalry as well as episodes of binocular fusion. 10.2307/1418496. Although both of these perceptual states clearly differ from exclusive visibility, they also do not correspond Alais, D., Cass, J., O’Shea, R. P., & Blake, R. (2010). to the perceptual switch. Binocular fusion is a default Visual sensitivity underlying changes in visual and different state of binocular vision (Wolfe, 1983). consciousness. Current Biology, 20, 1362–1367, doi: The piecemeal rivalry is the patchy appearance when 10.1016/j.cub.2010.06.015. some spatial regions are dominated by one eye whereas Blake, R., & Logothetis, N. K. (2002). Visual other regions are dominated by the other eye. It is competition. Nature Reviews. Neuroscience, 3, 13– more likely to occur for bigger visual displays (Blake et 21, doi:10.1038/nrn701. al., 1992; Kang, 2009; O’Shea, Sims, & Govan, 1997) Blake, R., O’Shea, R. P., & Mueller, T. J. (1992). and is reduced in the presence of additional grouping Spatial zones of binocular rivalry in central and factors such as rotation (Haynes, Deichmann, & Rees, peripheral vision. Visual Neuroscience, 8, 469–478, 2005). Accordingly, although one can talk about doi:10.1017/S0952523800004971. ‘‘mixed perception’’ with respect to the entire image, Blake, R., Westendorf, D., & Fox, R. (1990). Temporal individual patches are in exclusive visibility states. perturbations of binocular rivalry. Perception & Accordingly, although periods of nonexclusive visibil- Psychophysics, 48, 593–602, doi:10.3758/ ity may be long, they do not necessarily correspond to BF03211605. transient reversals of perceptual dominance postulated in current models of multistable perception (Laing & Brainard, D. H. (1997). The Psychophysics Toolbox. Chow, 2002; Moreno-Bote et al., 2011, 2007; Noest et Spatial Vision, 10, 433–436, doi:10.1163/ al., 2007). 156856897X00357. To summarize, perceptual reversals in SFM are Brascamp, J. W., & Blake, R. (2012). Inattention extremely brief, and it is for future research to abolishes binocular rivalry: Perceptual evidence. determine whether the same is true for perceptual Psychological Science, 23, 1159–1167, doi:10.1177/ switches in binocular rivalry and other multistable 0956797612440100. displays. However, it will be important to distinguish Brascamp, J. W., Blake, R., & Knapen, T. (2015). between perceptual switches and alternative perceptual Negligible fronto-parietal BOLD activity accom- states, such as piecemeal rivalry and binocular fusion. panying unreportable switches in bistable perception. Nature Neuroscience, 18, 1672–1678, doi:10. 1038/nn.4130. Conclusions Brascamp, J. W., van Ee, R., Noest, A. J., Jacobs, R. H. A. H., & van den Berg, A. V. (2006). The time course of binocular rivalry reveals a fundamental We report that induced perceptual reversals of role of noise. Journal of Vision, 6(11):8, 1244–1256, illusory rotation in SFM displays are very brief and doi:10.1167/6.11.8. [PubMed] [Article] occur in the absence of visual awareness. Brouwer, G. J., & van Ee, R. (2006). Endogenous Keywords: structure-from-motion, perceptual alterna- influences on perceptual bistability depend on tions, multistable perception, kinetic-depth effect, per- exogenous stimulus characteristics. Vision Research, ceptual reversals, visual awareness 46, 3393–3402, doi:10.1016/j.visres.2006.03.016.

Downloaded from jov.arvojournals.org on 09/23/2021 Journal of Vision (2016) 16(3):14, 1–16 Pastukhov & Klanke 15

Chong, S. C., Tadin, D., & Blake, R. (2005). patterns. Nature Neuroscience, 5, 605–609, doi:10. Endogenous attention prolongs dominance dura- 1038/nn851. tions in binocular rivalry. Journal of Vision, 5(11):6, Levelt, W. J. (1965). On binocular rivalry. Soesterberg, 1004–1012, doi:10.1167/5.11.6. [PubMed] [Article] the Netherlands: Institute for Perception RVO- Enns, J. T., & Di Lollo, V. (2000). What’s new in visual TNO. masking? Trends in Cognitive Sciences, 4, 345–352, Libet, B. (1999). How does conscious experience arise? doi:10.1016/S1364-6613(00)01520-5. The neural time factor. Brain Research Bulletin, 50, Faisal, A. A., Selen, L. P. J., & Wolpert, D. M. (2008). 339–340, doi:10.1016/S0361-9230(99)00143-4. Noise in the nervous system. Nature Reviews. Libet, B., Gleason, C. A., Wright, E. W., & Pearl, D. Neuroscience, 9, 292–303, doi:10.1038/nrn2258. K. (1983). Time of conscious intention to act in Fra¨ssle, S., Sommer, J., Jansen, A., Naber, M., & relation to onset of cerebral activity (readiness- Einha¨user, W. (2014). Binocular rivalry: Frontal potential): The unconscious initiation of a freely activity relates to introspection and action but not voluntary act. Brain: A Journal of Neurology, 106, to perception. Journal of Neuroscience, 34, 1738– 623–642, doi:10.1093/brain/106.3.623. 1747, doi:10.1523/JNEUROSCI.4403-13.2014. Martinez-Conde, S., Macknik, S. L., & Hubel, D. H. Gregory, R. L. (2009). Seeing through illusions (1st ed.). (2004). The role of fixational eye movements in New York: Oxford University Press. visual perception. Nature Reviews. Neuroscience, 5, Haynes, J.-D., Deichmann, R., & Rees, G. (2005). Eye- 229–240, doi:10.1038/nrn1348. specific effects of binocular rivalry in the human Metzger, W. (2009). Laws of seeing. Cambridge, MA: lateral geniculate nucleus. Nature, 438, 496–499, MIT Press. doi:10.1038/nature04169. Mitchell, J. F., Stoner, G. R., & Reynolds, J. H. (2004). Kang, M.-S. (2009). Size matters: A study of binocular Object-based attention determines dominance in rivalry dynamics. Journal of Vision, 9(1):17, 1–11, binocular rivalry. Nature, 429, 410–413, doi:10. doi:10.1167/9.1.17. [PubMed] [Article] 1038/nature02584. Kang, M.-S., & Blake, R. (2010). What causes Moreno-Bote, R., Knill, D. C., & Pouget, A. (2011). alternations in dominance during binocular rivalry? Bayesian sampling in visual perception. Proceed- Attention, Perception & Psychophysics, 72, 179–186, ings of the National Academy of Sciences, USA, 108, doi:10.3758/APP.72.1.179. 1–6, doi:10.1073/pnas.1101430108. Kim, Y.-J., Grabowecky, M., & Suzuki, S. (2006). Moreno-Bote, R., Rinzel, J., & Rubin, N. (2007). Stochastic resonance in binocular rivalry. Vision Noise-induced alternations in an attractor network Research, 46, 392–406, doi:10.1016/j.visres.2005.08. model of perceptual bistability. Journal of Neuro- 009. physiology, 98, 1125–1139, doi:10.1152/jn.00116. Knapen, T., Brascamp, J., Pearson, J., van Ee, R., & 2007. Blake, R. (2011). The role of frontal and parietal Naber, M., Fra¨ssle, S., & Einha¨user, W. (2011). brain areas in bistable perception. Journal of Perceptual rivalry: Reflexes reveal the gradual Neuroscience, 31, 10293–10301, doi:10.1523/ nature of visual awareness. PLoS One, 6, e20910, JNEUROSCI.1727-11.2011. doi:10.1371/journal.pone.0020910. Kornmeier, J., Ehm, W., Bigalke, H., & Bach, M. Nawrot, M., & Blake, R. (1989). Neural integration of (2007). Discontinuous presentation of ambiguous information specifying structure from stereopsis figures: How interstimulus-interval durations affect and motion. Science, 244, 716–718, doi:10.1126/ reversal dynamics and ERPs. Psychophysiology, 44, science.2717948. 552–560, doi:10.1111/j.1469-8986.2007.00525.x. Noest, A. J., van Ee, R., Nijs, M. M., & van Wezel, R. Laing, C. R., & Chow, C. C. (2002). A spiking neuron J. A. (2007). Percept-choice sequences driven by model for binocular rivalry. Journal of Computa- interrupted ambiguous stimuli: A low-level neural tional Neuroscience, 12, 39–53, doi:10.1023/ model. Journal of Vision, 7(8):10, 1–14, doi:10. A:1014942129705. 1167/7.8.10. [PubMed] [Article] Leopold, D. A., & Logothetis, N. K. (1999). Multi- Orbach, J., Ehrlich, D., & Heath, H. A. (1963). stable phenomena: Changing views in perception. Reversibility of the Necker cube. I. An examination Trends in Cognitive Sciences, 3, 254–264, doi:10. of the concept of ‘‘satiation of orientation.’’ 1016/S1364-6613(99)01332-7. Perceptual and Motor Skills, 17, 439–458, doi:10. Leopold, D. A., Wilke, M., Maier, A., & Logothetis, N. 2466/pms.1963.17.2.439. K. (2002). Stable perception of visually ambiguous Orbach, J., Ehrlich, D., & Vainstein, E. (1963).

Downloaded from jov.arvojournals.org on 09/23/2021 Journal of Vision (2016) 16(3):14, 1–16 Pastukhov & Klanke 16

Reversibility of the Necker cube: III. Effects of Ramachandran, V. S., & Anstis, S. M. (1983). interpolation on reversal rate of the cube presented Perceptual organization in moving patterns. Na- repetitively. Perceptual and Motor Skills, 17, 571– ture, 304, 529–531, doi:10.1038/304529a0. 582, doi:10.2466/pms.1963.17.2.571. Roeber, U., Veser, S., Schro¨ger, E., & O’Shea, R. P. O’Shea, R. P., Sims, A. J. H., & Govan, D. G. (1997). (2011). On the role of attention in binocular rivalry: The effect of spatial frequency and field size on the Electrophysiological evidence. PloS One, 6(7), spread of exclusive visibility in binocular rivalry. e22612, doi:10.1371/journal.pone.0022612. Vision Research, 37, 175–183, doi:10.1016/ Shpiro, A., Moreno-Bote, R., Rubin, N., & Rinzel, J. S0042-6989(96)00113-7. (2009). Balance between noise and adaptation in Pastukhov, A. (2015). Perception and the strongest competition models of perceptual bistability. Jour- sensory memory trace of multi-stable displays both nal of Computational Neuroscience, 27, 37–54, doi: form shortly after the stimulus onset. Attention, 10.1007/s10827-008-0125-3. Perception, & Psychophysics, 1–11, doi:10.3758/ Stonkute, S., Braun, J., & Pastukhov, A. (2012). The s13414-015-1004-4. role of attention in ambiguous reversals of struc- Pastukhov, A., & Braun, J. (2007). Perceptual reversals ture-from-motion. PloS One, 7, e37734, doi:10. need no prompting by attention. Journal of Vision, 1371/journal.pone.0037734. 7(10):5, 1–17, doi:10.1167/7.10.5. [PubMed] Treue, S., Andersen, R. A., Ando, H., & Hildreth, E. C. [Article] (1995). Structure-from-motion: Perceptual evidence Pastukhov, A., & Braun, J. (2011). Cumulative history for surface interpolation. Vision Research, 35, 139– quantifies the role of neural adaptation in multi- 148, doi:10.1016/0042-6989(94)E0069-W. stable perception. Journal of Vision, 11(10):12, 1– van Dam, L. C. J., & van Ee, R. (2006). Retinal image 10, doi:10.1167/11.10.12. [PubMed] [Article] shifts, but not eye movements per se, cause Pastukhov, A., & Braun, J. (2013). Structure-from- alternations in awareness during binocular rivalry. motion: Dissociating perception, neural persistence, Journal of Vision, 6(11):3, 1172–1179, doi:10.1167/ and sensory memory of illusory depth and illusory 6.11.3. [PubMed] [Article] rotation. Attention, Perception & Psychophysics, 75, van Ee, R. (2009). Stochastic variations in sensory 322–340, doi:10.3758/s13414-012-0390-0. awareness are driven by noisy neuronal adaptation: Pastukhov, A., Garc´ıa-Rodr´ıguez, P. E., Haenicke, J., Evidence from serial correlations in perceptual Guillamon, A., Deco, G., & Braun, J. (2013). bistability. Journal of the Optical Society of America Multi-stable perception balances stability and A, 26, 2612–2622, doi:10.1364/JOSAA.26.002612. sensitivity. Frontiers in Computational Neurosci- Weilnhammer, V. A., Ludwig, K., Hesselmann, G., & ence, 7, 17, doi:10.3389/fncom.2013.00017. Sterzer, P. (2013). Frontoparietal cortex mediates Pastukhov, A., Vivian-Griffiths, S., & Braun, J. (2015). perceptual transitions in bistable perception. Jour- Transformation priming helps to disambiguate nal of Neuroscience, 33, 16009–16015, doi:10.1523/ sudden changes of sensory inputs. Vision Research, JNEUROSCI.1418-13.2013. 116, 36–44, doi:10.1016/j.visres.2015.09.005. Wolfe, J. M. (1983). Influence of spatial frequency, Pastukhov, A., Vonau, V., & Braun, J. (2012). luminance, and duration on binocular rivalry and Believable change: Bistable reversals are governed abnormal fusion of briefly presented dichoptic by physical plausibility. Journal of Vision, 12(1):17, stimuli. Perception, 12, 447–456, doi:10.1068/ 1–16, doi:10.1167/12.1.17. [PubMed] [Article] p120447. Platonov, A., & Goossens, J. (2014). Eye dominance Wolfe, J. M. (1984). Reversing ocular dominance and alternations in binocular rivalry do not require suppression in a single flash. Vision Research, 24, visual awareness. Journal of Vision, 14(11):2, 1–17, 471–478, doi:10.1016/0042-6989(84)90044-0. doi:10.1167/14.11.2. [PubMed] [Article] Zhang, P., Jamison, K., Engel, S., He, B., & He, S. Prins, N., & Kingdom, F. A. A. (2009). Palamedes: (2011). Binocular rivalry requires visual attention. Matlab routines for analyzing psychophysical data. Neuron, 71, 362–369, doi:10.1016/j.neuron.2011.05. Retrieved from http://www.palamedestoolbox.org 035.

Downloaded from jov.arvojournals.org on 09/23/2021